Proposed here: 1. develop a general equilibrium prototype model of the poor high risk village or region as an open economy with risk from variable monsoon rains, crop, and human diseases, and variable prices; 2. derive under weak assumptions about technology and preferences optimal risk-sharing implications for age-sex adjusted per person consumption and individual labor supply, taking into account that family consumptions, not individual consumptions are measured in ICRISAT data, that family size and age compositions often vary considerably over time, that sampled households are a subset of the entire village or regional economy and that there is variable labor force participation due to sickness, disability, age, and temporary and permanent migration; 3. show the model's implications for final allocations allow the researcher to distinguish aggregate shocks, which are not insurable, from idiosyncratic shock, which potentially are insurable; 4. Conduct the appropriate statistical tests of the demographically-adjusted risk-sharing formulas: (a) age-sex adjusted per family member consumptions and individual leisures are determined by village or regional aggregates of these variables, only (b) having controlled for these aggregate variables, age-sex adjusted per family member consumptions and individual leisures should be immune from fluctuations in family crop outputs, profits from crop production, profits from all sources, wages, and full incomes and immune also from demographic variables such as age, household size, number of children and number of migrants (as not controlled for in the risk-bearing formulas); 5. quantify which, if any, economic or demographic shocks are not insured; similarly, verify if low-asset or low-income households are more likely to be uninsured; 6. document the extent of diversity across households in each village with respect to soil types, irrigation status, crop choice possibilities, and income sources. Thus quantify the extent to which households are doing the same thing and are subject to the same shocks to see if that explains observed comovements in consumptions. Similarly, derive household incomes after the buying and selling of physical stocks to see if households self-insure. Quantify the residual difference between income-after-self- insurance and consumption to measure the extent of community based transfers; and 7. classify and measure transfers by type (gifts, borrowing-lending) and by source (family,caste, other) to quantify the role of the immediate and extended family in the larger community context.